Bitcoin mining's post-halving economics have reached a structural inflection point. Hash price fell to $28/PH/s/day in February 2026 — a five-year low — according to CoinShares' Q1 2026 Bitcoin Mining Report. The weighted average cash cost to produce one bitcoin among publicly listed miners rose ...
"A number of sovereign funds are adding incrementally at $120,000, $100,000, establishing a longer position. This is not a trade." — Larry Fink, CEO, BlackRock
Bitcoin mining's post-halving economics have reached a structural inflection point. Hash price fell to $28/PH/s/day in February 2026 — a five-year low — according to CoinShares' Q1 2026 Bitcoin Mining Report. The weighted average cash cost to produce one bitcoin among publicly listed miners rose to approximately $79,995 in Q4 2025, while BTC trades at roughly $74,500 as of April 16, 2026. That arithmetic means the median public miner is producing bitcoin at a loss.
The result is a two-track industry. Efficient operators with sub-$0.05/kWh power and latest-generation hardware (≤12 J/TH) remain cash-flow positive. Everyone else is either shutting down rigs, selling reserves, or converting megawatts to AI and high-performance computing (HPC) infrastructure. More than $70 billion in AI/HPC contracts have been signed by former bitcoin-only miners since late 2024. CoinShares estimates that 15–20% of the global mining fleet is now unprofitable, and a difficulty adjustment scheduled for April 17, 2026, is projected to decrease difficulty by approximately 2.8%, from 138.97T to 135.14T — the latest in a series of downward adjustments reflecting sustained hashrate attrition.
Hash price — the daily revenue earned per petahash per second of deployed compute — peaked at approximately $63/PH/s/day in July 2025, according to CoinShares data. By November 2025, it had declined to $35–37/PH/s/day. In February 2026, it fell below $28/PH/s/day, establishing the lowest level since mid-2021.
The driver is straightforward: the April 2024 halving cut daily BTC issuance from roughly 900 BTC to approximately 437 BTC — a 51% reduction. Network hashrate, meanwhile, rose more than 30% in the subsequent twelve months, compressing per-unit revenue from both sides simultaneously.
Breakeven analysis varies by fleet composition:
| Operator Profile | Hardware Efficiency | Electricity Cost | BTC Breakeven Price | |---|---|---|---| | Tier 1 (latest gen) | ≤12 J/TH | ≤$0.05/kWh | ~$55,000 | | Tier 2 (mid-gen) | 15–20 J/TH | $0.05–0.07/kWh | ~$74,000 | | Tier 3 (legacy) | >30 J/TH | >$0.07/kWh | >$90,000 |
At Bitcoin's current price of approximately $74,500, Tier 2 operators are at or near cash-flow breakeven. Tier 3 operations are deeply underwater. The industry average "operating cash flow" breakeven sits near $77,000, according to data compiled by BydFi and corroborated by CoinShares estimates.
For the first time in six years, Bitcoin's network hashrate posted a Q1 decline, dropping approximately 4% year-to-date to hover near 940 EH/s as of April 7, 2026, according to multiple tracking sources. This follows a peak above 1.15 ZH/s in early 2026.
The decline is not uniform. On March 21, 2026, network difficulty fell 7.76% at a single adjustment — the second-largest downward adjustment in 2026 and one of the most significant in Bitcoin's history. According to CoinWarz and Newhedge data, the upcoming adjustment on April 17 is projected to reduce difficulty by a further 2.8%, bringing it from 138.97T to approximately 135.14T.
Over the past 90 days, cumulative difficulty has declined 5.12%. This pattern reflects what mining analysts have termed a "purge" of inefficient capacity: miners operating on power contracts above $0.07/kWh, running hardware older than Bitmain's S21 series, or located in jurisdictions with tightening regulatory environments have systematically gone offline.
CoinShares projects that total network hashrate could still reach 1.8 ZH/s by end of 2026 and 2 ZH/s by end of Q1 2027, assuming institutional buildout continues. The current dip, in this view, represents a temporary flush of marginal capacity rather than a secular decline.
The most consequential structural shift in the mining industry is the conversion of power capacity from bitcoin hashing to AI/HPC colocation. The economic logic is unambiguous: a 100 MW facility hosting AI workloads at $200/kW-month generates approximately $240 million per year in gross revenue. The same 100 MW facility mining bitcoin at current hash prices generates well under $100 million.
Major deals signed since late 2024 include:
The infrastructure cost differential is significant. Bitcoin mining buildout runs approximately $700K–$1M per MW. AI/HPC data center buildout costs $8M–$15M per MW, according to industry estimates. Miners with existing grid connections, substations, and cooling infrastructure can convert at a fraction of greenfield AI data center costs, giving them a structural advantage.
JPMorgan analysts estimate that miners with more than 500 MW of capacity could derive 30–50% of their revenue from AI by end of 2026, as hyperscalers including Microsoft and Google seek power sources outside congested utility grids.
CoinDesk reported on March 27, 2026, that miners are "becoming AI companies and selling their BTC to fund the transition" — a characterization supported by treasury data showing accelerating bitcoin liquidations among several publicly listed miners.
JPMorgan published a mining sector reset in early 2026, bifurcating the industry into AI-pivoting winners and mining-dependent laggards:
Upgraded:
Trimmed:
Analysts flagged share dilution as a rising headwind across the sector, with fully diluted share counts increasing by up to 30% at major miners.
Industry estimates project that by end of 2026, approximately 85% of global hashrate will be controlled by 12 or fewer entities — a mix of publicly traded companies and sovereign-backed operations. This represents a marked concentration from the more fragmented landscape of 2023–2024.
The survivors share common characteristics: sub-$0.05/kWh power, latest-generation ASICs (Bitmain S21 XP or equivalent at ≤12 J/TH), vertically integrated power procurement, and — increasingly — a parallel revenue stream from AI/HPC hosting.
Retail and mid-sized mining operations are functionally unviable at current economics. Legacy hardware with efficiency exceeding 30 J/TH is obsolete for mainnet bitcoin mining. The ROI timeline for new ASIC purchases has extended to approximately 1,000 days, according to CCN analysis — a figure that makes speculative entry prohibitive for all but the most capitalized operators.
Geographic distribution of hashrate is shifting. According to River data, 34 countries now control more than 0.1% of global hashrate, while 12 countries exceed 1%. Paraguay, Ethiopia, and Oman are each entering the top-10 hashrate rankings, reflecting a move toward jurisdictions with surplus renewable energy — predominantly hydroelectric power.
Energy-rich nations including Turkmenistan and Bhutan are using legislation or state-owned investment vehicles to convert mining into a mechanism for monetizing stranded energy and accumulating bitcoin as a strategic reserve asset. Twenty-three governments now hold bitcoin in some form, according to Cointribune reporting.
The concentration of hashrate among a small number of large entities and sovereign actors raises questions about mining centralization that extend beyond economics into network governance. When fewer than a dozen operators control the majority of block production, the practical distribution of consensus power narrows, regardless of the protocol's theoretical permissionlessness.
The AI pivot introduces a secondary concern. As miners redirect power from bitcoin hashing to GPU colocation, bitcoin's security budget — the total economic incentive sustaining consensus — depends on a shrinking pool of dedicated mining operators. The March 21, 2026, difficulty drop of 7.76% represented a real, if temporary, reduction in the computational cost of attacking the network.
Techi.com reported in 2026 that the miner-to-AI pivot poses a "Bitcoin security risk" if hashrate growth fails to keep pace with network value. CoinShares' hashrate projections assume continued institutional buildout will offset the loss of marginal miners, but the transition period creates a window of reduced security margin relative to network capitalization.
The bitcoin mining industry in April 2026 is undergoing a forced restructuring driven by post-halving revenue compression and the superior economics of AI data center hosting. The hash price collapse has eliminated marginal operators, concentrated hashrate among a dozen well-capitalized entities, and catalyzed a $70 billion+ reallocation of power infrastructure toward GPU workloads.
This is not a cyclical downturn. The structural gap between AI hosting revenue ($240M/year per 100 MW) and bitcoin mining revenue (<$100M/year per 100 MW) creates a permanent economic incentive for miners to redirect capacity. The question is no longer whether bitcoin mining becomes a secondary function of large-scale power companies — it is how quickly, and what the implications are for Bitcoin's consensus security as dedicated hashrate becomes a smaller share of these operators' total revenue.
CoinShares' projection of 1.8 ZH/s by end of 2026 assumes the industry resolves this tension. Whether that assumption holds depends on bitcoin's price trajectory and the degree to which institutional capital views mining as a strategic reserve function rather than a standalone business.